ClawHub
Back to skill

Security audit

exam-study-guide-predictor

Security checks for vulnerabilities and agentic risk

Overview

This skill creates exam study guides from user-provided course materials, with disclosed and proportionate file output behavior.

Before installing, expect the skill to process any course files you provide and to create output files in your user outputs folder. Avoid giving it unrelated private documents, and ask for your preferred output language if English is not suitable.

SkillSpector

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The skill’s trigger language is broad enough to catch generic study-related requests such as 'help me study' or 'make me a revision guide,' which can cause the agent to invoke this specialized workflow in contexts the user did not intend. That increases the chance of over-collection of educational materials, mismatched task execution, and accidental steering into exam-prediction behavior when a simpler tutoring or summarization response would have been more appropriate.

Natural-Language Policy Violations

Medium
Confidence
82% confidence
Finding
The instruction 'Output in English, but read Arabic' forces a single output language without user choice, which can cause the system to present sensitive academic guidance in a language the user may not prefer or fully understand. In multilingual settings this can reduce accuracy, accessibility, and user control, especially when source materials and exam cues are in Arabic and nuanced wording matters.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill instructs the agent to save generated artifacts directly into `/mnt/user-data/outputs/` and present them, without requiring explicit user confirmation or warning that files will be created. In an agent environment, silent file creation is a real side effect: it can surprise users, create unwanted persistent artifacts, and normalize write operations to user storage without consent.

Static analysis

No suspicious patterns detected.